MODELING THE DYNAMICS OF UKRAINE’S GROSS DOMESTIC PRODUCT IN THE CONTEXT OF CONTEMPORARY GEOPOLITICAL CHALLENGES

Abstract

The article analyzes the dynamics of Ukraine's gross domestic product (GDP) from 2013 to 2024. Correlation and regression analysis methods were used to identify significant and linearly independent factors influencing GDP during the specified period. Two multifactor nonlinear regression models of Ukraine's GDP were constructed, one of which is dynamic and includes a correction factor reflecting the combined impact of new factors that emerged during the full-scale war but are not explicitly included in the model as explanatory variables. According to the constructed models, factor X1 (capital investment) and factor X3 (real effective exchange rate of the hryvnia) had a positive effect on Ukraine's GDP (Y). A 1% increase in investment led to a 0.83% increase in GDP, while a 1% increase in the real effective exchange rate of the hryvnia resulted in a 0.29% increase in GDP. Factor X2 (population), when decreasing, leads to an increase in the dependent variable (Y). Such a relationship between X2 and Y may indicate structural changes in the economy, including a transition from labor-intensive to capital-intensive and high-tech industries, and may also reflect increased production in the defense sector during the war. A forecast for 2025 was generated using the constructed models. A comparison of the relative forecast errors obtained from different models favors the model that incorporates the correction factor. The first model (essentially a static multivariate nonlinear regression model) is characterized by a high coefficient of determination (R² = 0.96117), indicating a good fit to the initial data and its potential usefulness for forecasting; however, it also shows a relatively large forecast error for 2025 (-17.24%). The second model (which includes a correction factor) has a slightly lower coefficient of determination (R² = 0.9193); however, its forecast error is significantly smaller (-2.095%). Thus, the correction factor α(t) introduces a dynamic component into the model, improving the accuracy of GDP forecasts. Prospects for further research include the development of a clear classification of factors, assessment of their impact on GDP under current conditions, and a more differentiated representation of these effects in relevant economic and mathematical models.